loss_.backward()
optimizer.step()
train_l_sum += loss_.cpu().item()
train_acc_sum += (y_hat.argmax(dim=1) == y).sum().cpu().item()
n += y.shape[0]
batch_count += 1
test_acc = evaluate_accuracy(test_iter, net)
print('epoch %d, loss %.4f, train acc %.3f, test acc %.3f,\
time %.1f sec' %
(epoch + 1, train_l_sum / batch_count, train_acc_sum / n,
test_acc, time.time() - start))
lr, num_epochs = 0.01, 5
optimizer = torch.optim.Adam(filter(lambda p: p.requires_grad,
net.parameters()),
lr=lr)
loss = nn.CrossEntropyLoss()
train(train_iter, test_iter, net, loss, optimizer, device, num_epochs)
def predict_sentiment(net, vocab, sentence):
"""sentence是词语的列表"""
device = list(net.parameters())[0].device
sentence = torch.tensor([vocab.stoi[word] for word in sentence],
device=device)
label = torch.argmax(net(sentence.view((1, -1))), dim=1)
return 'positive' if label.item() == 1 else 'negative'
def main():
print("索引化后的训练数据:", len(train_features))
print("索引化后的测试数据:", len(test_features))
print("训练集标签:", len(labels))
from model import generator, BiRNN, train, evaluate_accuracy
from torch import nn, optim
import torch
from sklearn.model_selection import train_test_split
x_train, x_test, y_train, y_test = train_test_split(np.array(train_features),
np.array(labels),
test_size=0.2)
x_train = torch.tensor(x_train, dtype=torch.long)
y_train = torch.tensor(y_train)
x_test = torch.tensor(x_test, dtype=torch.long)
y_test = torch.tensor(y_test)
test_features = torch.tensor(np.array(test_features), dtype=torch.long)
embeddings_matrix = torch.tensor(embeddings_matrix)
net = BiRNN(word_index, 100, 100, 2)
net.embedding.weight.data.copy_(embeddings_matrix)
net.embedding.weight.requires_grad = False
loss = nn.CrossEntropyLoss()
optimizer = optim.Adam(filter(lambda p: p.requires_grad, net.parameters()),
lr=0.0001)
num_epochs = 10
train_iter = generator(x_train, y_train, 10, train=True)
test_iter = generator(x_test, y_test, 10, train=False)
train(train_iter, test_iter, net, loss, optimizer, num_epochs)